Intelligent workload balance control of the assembly process considering condition-based maintenance
Introduction
Assembly line balancing problem (ALBP) is key to the efficiency of the assembly process. However, the assembly process is prone to disruptions. Disruptions associated with workstations, buffers or raw materials can affect the assembly process negatively, and the adverse impact can propagate along the assembly line. Thus, the assembly process should be monitored and controlled in real time to smooth the workflow. Industry 4.0, the fourth industrial revolution, brings new possibilities for the assembly systems. Internet of Things (IoT), as the basic premise for the implementation of Industry 4.0 [1], has a significant influence on the developments of smart workshop [2]. Cyber-Physical Systems (CPS), which is another key technology of Industry 4.0, can manage the integration of data collected from a factory, and enables information to be monitored and synchronized between the physical world and the cyber environment [3]. With the real-time information obtained, the production process can be optimized and re-optimized, thus, intelligent manufacturing, which will bring revolutionary changes [4], is possible. Smart systems such as smart robotic mobile fulfillment system [5], warehouse management system for smart logistics [6], smart product-service systems [7], [8], smart active maintenance system [4], smart suite for smart factory towards re-industrialization [9] and smart home system [10] have been proposed based on the advanced technologies of Industry 4.0.
With the development of information and communication technologies (ICT), information technology is embedded in normal products and transforms them to smart, connected products so that information can be generated [11]. IoT brings internet to all kinds of devices so that they can not only collect and send information but also receive information. Additionally, IoT can significantly affect the tracking applications [2], therefore, the work-in-progress during the production process can be tracked more easily and accurately. Due to the digital transformation enabled by ICT, physical products can be digitized in the virtual space and interconnected [12]. As a result, a workstation of an assembly line can ‘communicate’ with its upstream and downstream workstations and buffers to get access to both local information and global information. Then, information barriers between different parts of an assembly line can be broken, and centralized and decentralized decision-making are possible. Robotic process automation (RPA), a new technology that consists of software agents that mimic the manual routine decisions via various computer applications, has attracted attention in Industry and academia [13]. It is enabled by the advanced information technologies, and brings new possibilities to the assembly process: workstations can work autonomously with each other to satisfy the demand, and can collaborate with each other to deal with unexpected events and decrease work-in-progress. Since the benefits brought by the advanced technologies can be fused by the automation with analytics and decision-making via tools of artificial intelligence, traditional automation of the assembly process can be improved to be intelligent automation [14]. Therefore, an intelligent control system of the assembly process enabled by the latest technologies of Industry 4.0 is possible, and it is necessary to decrease the negative impacts of unexpected events and increase the line efficiency.
Machine degradation can affect the duration of job processing, and will negatively affect the workload balance of an assembly line. Thus, the degradation process and machine state should be monitored so that workloads of workstations can be balanced with the real-time processing abilities of workstations. With the development of sensor technologies, it is possible to monitor the machine degradation process, and the information collected can be used to build and optimize the schedule of condition-based maintenance, which is to implement maintenance operations based on the online measurements of machine degradation level [15]. Besides, IoT enables real-time information collection to be possible, and helps establish efficient maintenance strategies at low cost with the related information [10]. The findings of Ghaleb et al. [16] indicated that significantly increase savings could be obtained when the accurate information of machines’ degradation process were incorporated in the condition-based maintenance strategies. Thus, smart machines can send their working status to a central cloud-based “manager” [17], and real-time degradation information can be levered to implement condition-based maintenance and improve the health state of each machine at the right time.
Some researchers explored the production scheduling problems and modeled the impact of deterioration effect on job processing time with a linear or non-linear function of start time (e.g. [18], [19]) or the position of the job (e.g. [20], [21]). However, the link between the extent of machine degradation and production re-scheduling was not discussed by these studies. With respect to assembly line re-balancing, to react quickly to the disruptions to the assembly process, Belassiria et al. [22] and Girit and Azizoğlu [23] explored the assembly line re-balancing solution with methods based on genetic algorithm and both exact and tabu search algorithm, respectively. Meanwhile, to respond quickly to changes, Moghaddam et al. [24] utilized reconfigurable manufacturing systems with modular reconfigurable machine tools to adjust production capacity of the system. However, these researches focused on searching the re-optimization solutions with the assumption that the re-balancing decision has been made already, and when to re-balance the assembly line was still not examined.
The link between the extent of disruptions and the decision-making process of assembly line re-balancing is examined by only a few researches. Suwa [25] and Valledor et al. [26] proposed the periodic re-scheduling policy, and whether to implement re-scheduling was determined based on the condition at the predetermine inspection times. However, the effectiveness of such policies would inevitably depend on the predetermined inspection interval, and the real-time decision-making was not possible. In our previous work [27], [28], a fuzzy control system was proposed to determine when to re-balance the workloads and manage the inventory level of work-in-progress. Nevertheless, it was assumed that no maintenance activities would be conducted until machines of an assembly line broke down. Therefore, it is necessary to build a real-time control system and develop a link between the extent of disruptions and the re-balancing decision, and real-time information of the machine degradation process should be utilized to implement condition-based maintenance.
Due to the randomness and non-linearity caused by unpredictable disruptions, accurate analysis of the assembly line is difficult. However, fuzzy controllers provide an efficient architecture to incorporate the linguistic information from the knowledge of experts to the final automated decisions. Thus, fuzzy control theory is chosen to design the control system of the assembly process, and is used to determine the trigger point of assembly line re-balancing in a novel way. To fill the above research gaps, this study integrates the fuzzy logic principles and the assembly line monitoring and control mechanism, and enables the effective decision-making process of how to manage the inventory level of the work-in-progress and when to re-balance the assembly line, which is of great importance to the efficiency of assembly lines but always ignored in the current literature. During the control process, fluctuations of processing abilities of machines due to machine degradation and condition-based maintenance are considered. A fuzzy control system is proposed to analyze the real-time information collected from the assembly process and balance the workloads by not only adjusting the production rates of workstations but also re-balancing the workloads by task re-assignments when necessary. Two kinds of assembly line re-balancing strategies are used so that two levels of modifications of the initial assembly plan can be implemented to the assembly line. The machine degradation process is assumed to be monitored continuously in real time so that the changing processing abilities of workstations can be considered in the decision-making process. This study explores the smart balance control of an assembly line with the consideration of the information brought by the advanced technologies in Industry 4.0. The research findings will shed light on the intelligent control of the assembly process and contribute to the smart manufacturing theory.
The remaining sections of this paper are organized as follows. In Section 2, literature on production scheduling considering the deterioration effect and the real-time control of the production process is reviewed. Section 3 shows the problem definition and the proposed fuzzy control system. The results of the numerical experiments are shown in Section 4. Section 5 presents the conclusions of this study.
Section snippets
Production scheduling considering the deterioration effect
In the traditional production scheduling problems, it is assumed that the processing time of a job is constant. However, due to the deterioration effect, the duration of job processing increases with the increase of the total operation time; The duration decreases if the learning effect is considered.
The impact of machine degradation on job processing is modeled by a linear increasing function of the job’s starting time by some researchers. For example, Woo and Kim [29] explored a parallel
Problem statement
Fig. 1 shows the structure of the assembly line considered in this study. denotes buffer . Buffers to have finite capacity, and and have infinite capacity since they are used to store raw materials and the finished products.
Machine degradation can affect the processing time of a job [16]. Thus, information of machine degradation is useful to retain the workload balance of an assembly line. It is assumed that machine deterioration is monitored continuously. Let and be
A real-time assembly line balancing method
With the consideration of machine degradation and condition-based maintenance, processing abilities of machines fluctuate with the changing health state. Thus, it is necessary to monitor the assembly process and balance the workloads of workstations based on the real-time information of machines’ processing abilities. In this study, a fuzzy control system is proposed to monitor and adjust the workload of workstations in real time. Assembly line re-balancing will be conducted to reduce the
Numerical experiments
The task times and precedence relationship of the assembly line are initialized by an instance of assembly line balancing problem: KILBRID (45 tasks). The processing times of tasks and precedence relationship are in the SALBP data sets shown on https://assembly-line-balancing.de/salbp/benchmark-data-sets-1993/. In order to differentiate the output of different assembly lines, the processing times used in the numerical experiments are set to be 100 times smaller than the initial ones.
Since the
Conclusions
In this paper, machine degradation process is modeled by the gamma process so that the degradation level is strictly increasing. The impact of machine degradation on the duration of task processing is considered. The degradation process is divided into several stages, and the processing abilities of machines vary with their health state. In order to balance the workloads of workstations based on the real-time processing abilities associated with the health state of machines, a fuzzy control
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This research is funded by the Laboratory for Artificial Intelligence in Design (Project Code: RP2-2), Hong Kong Special Administrative Region, Hong Kong, China. The authors also would like to thank the support of Department of Industrial and Systems Engineering, The Hong Kong Polytechnic University, Hong Kong, China.
References (50)
- et al.
IoT edge computing-enabled collaborative tracking system for manufacturing resources in industrial park
Adv. Eng. Inform.
(2020) - et al.
Smart robotic mobile fulfillment system with dynamic conflict-free strategies considering cyber-physical integration
Adv. Eng. Inform.
(2019) - et al.
Edge-cloud orchestration driven industrial smart product-service systems solution design based on CPS and IIoT
Adv. Eng. Inform.
(2019) - et al.
IoT-enabled smart appliances under industry 4.0: A case study
Adv. Eng. Inform.
(2020) - et al.
A novel data-driven graph-based requirement elicitation framework in the smart product-service system context
Adv. Eng. Inform.
(2019) - et al.
A survey of smart product-service systems: Key aspects, challenges and future perspectives
Adv. Eng. Inform.
(2019) - et al.
Robotic process automation: contemporary themes and challenges
Comput. Ind.
(2020) - et al.
Intelligent automation of design and manufacturing in machine tools using an open architecture motion controller
J. Manuf. Syst.
(2013) - et al.
Joint optimization of condition-based maintenance and production lot-sizing
Eur. J. Oper. Res.
(2016) - et al.
Scheduling of unrelated parallel machines considering sequence-related setup time, start time-dependent deterioration, position-dependent learning and power consumption minimization
J. Clean Prod.
(2020)
A note: Multi-machine scheduling with general position-based deterioration to minimize total load
Int. J. Prod. Econ.
A new when-to-schedule policy in online scheduling based on cumulative task delays
Int. J. Prod. Econ.
Smart control of the assembly process with a fuzzy control system in the context of Industry 4.0
Adv. Eng. Inform.
Matheuristic approaches for parallel machine scheduling problem with time-dependent deterioration and multiple rate-modifying activities
Comput. Oper. Res.
Parallel-machine scheduling with linear deteriorating jobs and preventive maintenance activities under a potential machine disruption
Comput. Ind. Eng.
Combining time and position dependent effects on a single machine subject to rate-modifying activities
Omega-Int. J. Mange. S.
Due date assignment and multitasking scheduling with deterioration effect and efficiency promotion
Comput. Ind. Eng.
Scheduling jobs with step-deterioration; minimizing makespan on a single-and multi-machine
Comput. Ind. Eng.
Fuzzy work-in-process inventory control of unreliable manufacturing systems
Inf. Sci.
Joint optimization of monitoring quality and replacement decisions in condition-based maintenance
Reliab. Eng. Syst. Saf.
Condition-based maintenance under performance-based contracting
Comput. Ind. Eng.
Modeling age-based maintenance strategies with minimal repairs for systems subject to competing failure modes due to degradation and shocks
Eur. J. Oper. Res.
Optimal condition-based maintenance policy with delay for systems subject to competing failures under continuous monitoring
Comput. Ind. Eng.
A prediction method for the real-time remaining useful life of wind turbine bearings based on the Wiener process
Renew. Energy
A methodology for solving single-model, stochastic assembly line balancing problem
Omega-Int. J. Manage. Sci.
Cited by (7)
Reliability improvement on assembly accuracy with maximum out-of-tolerance probability analysis and prior precise repair optimization
2023, Advanced Engineering InformaticsIndustrial internet of things-driven storage location assignment and order picking in a resource synchronization and sharing-based robotic mobile fulfillment system
2022, Advanced Engineering InformaticsCitation Excerpt :SCADA system is an assemblage of both hardware and software components that are event-driven and allow local and remote supervision and control. Multiple tasks need to monitor the suppliers’ manufacturing [100–102]. The SCADA system is integrated with the ERP system, including remote monitoring and controlling based on the cloud-based SCADA system.
Editorial Notes: Emerging intelligent automation and optimisation methods for adaptive decision making
2022, Advanced Engineering InformaticsEquipment electrocardiogram (EECG): making intelligent production line more robust
2023, Journal of Intelligent Manufacturing